I have income as a continuous variable and test scores as a continuous variable. I know that there is a correlation between income and test scores, but what I really want to find is the optimal income or income rank for predicting high test scores. What analysis shouldI use? Is kind of latent class analysis?
first, you need to check whether the correlationship between income and test score, using Pearson Correlation.
it may use variable as categories data, Categ. A give more than 5000$
Categ. B gives 4000 upto less than 5000, and so on
I suggest breaking the data set into ten income deciles and running Pearson r for each separately. The one with the strongest correlation would be the best for predicting high test scores. This type of analysis will also indicate whether or not there is a curvilinear relationship between income and test scores. Because of the likely negative skew for income, the correlation in the highest income decile might even be negative.
A stats guy at Georgia walked me through this in a secondary analysis of the 2013 Survey of Consumer Finances.
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis